(470b) Streamlined Yeast Cell Reactors with Residence Time Control to Engineer and Profile Proteases

Proteases, enzymes responsible for cleaving proteins through the hydrolysis of peptide bonds, make up approximately 2% of the proteome (1). Over the last few decades, there has been a growing interest in proteases with tailored substrate specificities for proteome editing. Proteases with bespoke substrate specificity could be used as therapeutic agents by degrading disease-associated proteins, ushering in a new wave of precision proteome targeting for disease treatment.

Despite their therapeutic potential, many protease-based drugs face challenges advancing beyond clinical trials, primarily due to their broad substrate specificity (2). Successfully altering protease substrate specificity remains challenging. Often, when proteases are evolved to cleave a desired substrate without a counterselection substrate, the final protease product retains high catalytic activity on the native substrate (3). To address these challenges, the Yeast Endoplasmic Reticulum (ER) Sequestration and Screening (YESS) system emerged as a powerful tool for quantifying protease activity and selecting desired variants via yeast surface display (YSD) (4). The YESS system enables precise control of enzyme: substrate stoichiometries and reaction rates through its transcriptional (promoters of different strengths) and post-translational nodes (ER retention signal (ERS) strengths). In the present iteration of the YESS system, however, the counterselection substrate (CS) and selection substrate (SS) sequences are located on a single polypeptide, making it impossible to change the stoichiometric ratios of CS and SS independently. This limited functionality makes it impossible to engineer protease substrate specificity under high ratios of CS to SS, which are ideal for guiding evolution campaigns.

Here, we present a modified version of YESS that exploits the correlation between ER retention sequence (ERS) strength and ER residence time. By integrating two substrate cassettes— one with a strong ERS and another with no ERS—into the yeast chromosome, we bias protease-substrate exposure (Figure 1A,B). We showed that using this system a substrate cassette with no ERS results in faster transport to the yeast surface than a substrate cassette with a strong WEHDEL ERS (Figure 1C). We also used this version to test the orthogonality of a TEV protease variant, TEVEp, engineered in the original version of YESS, to cleave after a P1 glutamate and away from the canonical P1 glutamine (4). TEVEp, which shows a three-fold preference in catalytic efficiency for ENLYFES over ENLYFQS, failed to show this presence in our new system (Figure 1D).

Furthermore, using the dual display system, we can screen through a library of protease variants to isolate those highly specific for cleaving only the SS. By pairing the SS with no ERS and the CS with a strong ERS, proteases must avoid cleaving the CS during extended exposure while efficiently cleaving the SS, even in brief interactions. Demonstrated, by the evolution of a TEVEp variant with higher specificity for glutamate at P1. The customizability of the ER residence time dual display system enables enzyme substrate specificity engineering under various stringencies. This highly modular tool will enable us to hijack a variety of proteases to leverage their immense physiological impact to improve human health. Moreover, enhanced protease substrate specificity enables precise peptide cleavage, thereby improving the effectiveness of biotherapeutics, commercial products, and driving advancements in proteomics and synthetic biology.

References: (1) Lim, M., et al., Bioorg. Med. Chem. 2009. (2) Rudzińska, M.; et al., Drug Des. Devel. Ther. 2021. (3) Varadarajan, N., et al., Proc. Natl. Acad. Sci. 2013. (4) Yi, L., et al., Proc. Natl. Acad. Sci. 2013.